Introduction

Supination injury of the ankle is one of the most common injuries in the general population and elite athletes. The majority of these injuries resolve with rehabilitation, but chronic sequelae can occur [1]. Soft tissue and osseous impingement syndromes are an important cause of ankle dysfunction in athletes.There are other causes, however, including mechanical instability and osteochondral lesions [1]. Although impingement syndromes may coexist with other causes of chronic pain they may still be the predominating clinical feature [2, 3].

Osseous and soft tissue impingement syndromes have been described in the anterolateral, anterior, anteromedial, posterior and posteromedial ankle [2–8]. Posteromedial impingement is the least-described ankle impingement syndrome, with the exact aetiology and characteristic imaging features not completely defined [3, 7, 9]. The condition is thought to develop after compression of the posteromedial tibiotalar capsule and posterior fibres of the tibiotalar ligament (PTTL) between the talus and medial malleolus during an supination injury (Fig. 1) [3]. It is believed that subsequent fibrosis and thickening of the contused PTTL and posteromedial capsule leads to painful impingement between the medial wall of the talus and posterior margin of the medial malleolus [3]. Posteromedial tenderness on inversion with the ankle in plantar flexion is an important clinical finding which differentiates pain originating from tibialis posterior abnormality [3].

Fig. 1
figure 1

Anatomy of the posteromedial ankle and posterior tibiotalar ligament. Line drawing at the level of the talus (T) and distal medial malleolus (M) shows the deep tibiotalar ligament (white arrow) with the more superficial posterior tibiotalar ligament and capsular tissue (black arrows) lying deep to tibialis posterior (asterisk)

Full size image

This report aims to describe the use of MR imaging and evaluate the efficacy of ultrasound-guided steroid injection in the diagnosis and management of nine elite athletes with subacute clinical posteromedial impingement of the ankle developing after an initial supination injury.

Subjects and methods

After institutional ethics committee approval a retrospective analysis of nine consecutive elite athletes who underwent MR imaging and ultrasound-guided injection for subacute clinical posteromedial impingement was undertaken. Eight athletes were male and one female, with a median age of 29 years (range 17–40 years). Sports played were professional soccer (n=5), rugby (n=2), cricket (n=1) and ballet (n=1).

Clinical assessment

All of the athletes were referred by team clinicians and had a history of persisting symptoms following an initial supination injury. Despite rehabilitation with resolution of anterior and lateral symptoms all developed resistant posteromedial pain 3–4 weeks after the injury, limiting athletic activity. All denied previous posterior ankle pain.

Clinical examination demonstrated posteromedial tibiotalar tenderness and soft tissue thickening anterior to and not involving the Achilles tendon. Symptoms and point tenderness were exacerbated by inversion of the ankle in plantar flexion. There was no clinical evidence of posteromedial tenosynovitis or mechanical instability.

MR imaging evaluation

All athletes underwent MR imaging of the ankle performed on a 1.5-T system (Intera; Philips, Bothell, WA). For each examination axial PD-weighted (TR/TE 2800/15), sagittal T1-weighted conventional spin echo (TR/TE 400/13), and axial, sagittal and coronal T2-weighted fast spin echo fat-suppressed (TR/TE 3540/103) images were obtained. Axial and sagittal T1 weighted conventional spin echo fat suppressed (TR/TE 675/20) gadolinium (Magnevist; Schering-Plough, Madison, NJ)-enhanced images were obtained in all athletes. All sequences were performed using a head coil (Philips, Bothell, WA) with a slice thickness of 3 mm, field of view 127–160 mm and matrix of 256×256. Gadolinium-enhanced sequences were added to the imaging protocol to highlight enhancement within the capsular and pericapsular soft tissues [10, 11]. A formal comparison of imaging sequences was not performed.

MR imaging examinations were retrospectively reviewed by two experienced musculoskeletal radiologists along with MR imaging examinations from six professional athletes with clinical posterolateral impingement [all male, median age 24 years, range 23–30 years, sports: soccer (n=4), cricket (n=1), squash (n=1)].

The radiologists were blinded to the clinical details and proportion of cases. The presence and distribution of capsulitis/ synovitis, bone marrow oedema, os trigonum (and, if present, synchondrosis disruption), and collateral ligament, osteochondral and tendon abnormality was scored for each examination as normal, mild, moderate or severe by consensus. The posterior ankle was evaluated in three areas for synovitis: posteromedial (posterior to PTTL and medial to medial talar process), posterior (posterior to both talar processes) and posterolateral (posterior to posterior talofibular ligament and lateral to lateral talar process). Displacement and encasement of the tendons by posteromedial soft tissue thickening was also assessed (absent or present), as these imaging features had been identified in a previous series [7].

Ultrasound evaluation and guided injection

All nine athletes subsequently underwent ultrasound (Antares; Siemens Ultrasound, Mountain View, CA) by an experienced radiologist using a 13–5 MHz transducer. The posteromedial capsular soft tissues, PTTL and posteromedial tendons were evaluated for abnormality at rest and dynamically including Doppler assessment (for increased vascularity, recorded as absent or present). Evaluation of the posteromedial ankle was performed with the transducer in the transverse or oblique axial plane to visualise the posterior medial malleolus, tibialis posterior, flexor digitorum longus and the capsular tissues deep to the tendons. The dynamic part of the examination included plantar- and dorsiflexion of the ankle to evaluate movement of any capsular thickening and associated adherence to the overlying tendon sheaths.

Ultrasound-guided injection was then performed into the posteromedial capsular abnormality. Median time from MR imaging to ultrasound was 3 days (range 1–7 days). After choosing an oblique axial transducer position to visualise the posteromedial tissues and using standard aseptic technique a 20- or 23-gauge needle was directly guided to the abnormal area of the capsule. After dry needling (multiple passes without injection) throughout the abnormal tissues, infiltration of the abnormal tissues with 40 mg triancinolone acetonide (Kenalog; Bristol-Myers Squibb, NY) and 3 ml 0.5% bupivocaine (Marcain; AstraZeneca, London, UK) was performed.

Follow-up

After injection the athlete could leave the department within 30 min and rested for 48–72 h before commencing light training with gradually increasing intensity over the next 10–14 days.

Athletes were followed up by telephone interview at 4 weeks and then at 6-monthly intervals. Time to return to previous professional activity and any further interventions or surgical procedures were recorded.

Results

MR imaging evaluation

A comparison of MR imaging findings for the posteromedial and posterolateral groups is shown in Table 1.

Table 1 Incidence of MR imaging abnormality (mild, moderate or severe) for athletes with posteromedial and posterolateral pain
Full size table

Posteromedial capsular thickening causing tendon displacement was seen only in athletes with posteromedial impingement (7/9) and was graded as mild in five cases, moderate in one case (Fig. 2) and severe in the remaining case. No tendon encasement was seen. Posteromedial synovitis was present in all athletes with posteromedial impingement (mild=4, moderate=2, severe=3) (Figs. 2 and 3). However, posterior and posterolateral synovitis was also seen in these athletes and scored the same (n=3), less (n=3) and more severe (n=3) than the posteromedial changes.

Fig. 2
figure 2

A 23-year-old professional soccer player with clinical posteromedial impingement of the right ankle. a Axial PD-weighted MR image (TR/TE 2903/15, echo train length 8) shows loss of striation of the tibiotalar ligament (arrowheads) and posterior and posterolateral capsular fluid (arrow). b Axial T1-weighted spin echo fat-suppressed gadolinium enhanced MR image (TR/TE 456/12) shows enhancing posteromedial capsular thickening (arrowheads) displacing tibialis posterior. Note anterolateral talar oedema (asterisk). c Axial oblique ultrasound image of the posteromedial talus (Ta) and medial malleolus (M) obtained during injection shows predominantly hypoechoic capsular thickening (asterisk) with a hyperechoic nodule (arrow) displacing tibialis posterior (T) and the needle (arrowheads) placed prior to infiltration and injection. d Same level as c during injection with hyperechoic steroid and anaesthetic injected into the loose capsular thickening, causing further displacement of tibialis posterior

Full size image
Fig. 3
figure 3

A 26-year-old professional soccer player with clinical posteromedial impingement of the right ankle. a Axial T2-weighted fat-suppressed MR image (TR/TE 2000/90, echo train length 14) shows normal striation of the tibiotalar ligament (arrow), posterolateral capsular fluid (small arrowheads) and mild posteromedial synovitis (large arrowhead). b Axial T1-weighted spin echo fat-suppressed gadolinium-enhanced MR image (TR/TE 456/12) shows enhancing mild posteromedial capsular thickening (arrowheads) with no displacement of tibialis posterior. c Axial oblique ultrasound image of the posteromedial talus (Ta) and medial malleolus (M) shows hypoechoic and hyperechoic capsular thickening (asterisk) approaching but not displacing tibialis posterior (TP) or flexor digitorum longus (FDL)

Full size image

In contrast mild posteromedial synovitis was present in two of the posterolateral athletes with no evidence of tendon displacement. All posterolateral athletes had moderate (n=3) or severe (n=3) posterolateral synovitis.

Bone marrow oedema was present in the posterolateral tibia (n=2), anterior tibia (n=1), hypertrophied medial talar process (n=1), anterior talus (n=2), posterolateral talus (n=1) and os trigonum (n=1), with no specific distribution in either group. The athlete with the hypertrophied medial talar process had clinical posteromedial impingement (Fig. 4). An os trigonum was present in four athletes, equally divided between the two groups, with disruption of the synchondrosis present in one athlete with posterolateral symptoms.

Fig. 4
figure 4

A 21-year-old professional rugby player with clinical posteromedial impingement of the right ankle. a Axial CT image shows the lateral talar process, flexor hallucis longus (white arrowhead) and an enlarged irregular medial talar process (black arrowheads). b Axial T2-weighted fat-suppressed MR image (TR/TE 2000/90, echo train length 14) shows the oedematous medial process of the talus (asterisk) and posteromedial synovitis (arrow) posterior to the tibiotalar ligament (large arrowhead) extending to tibialis posterior and flexor digitorum longus (small arrowheads). c Axial oblique ultrasound image with extended field of view shows the medial malleolus (M), tibialis posterior (TP) and the irregular posteromedial process (TPr) of the talus. Capsular thickening (arrowheads) is present on the superficial surface of the process, displacing flexor digitorum longus (arrow). d Sagittal sonogram at the level of flexor digitorum longus (arrowheads) shows the capsular thickening (asterisk) and irregular medial process (arrows). e Sagittal sonogram shows the needle (arrowheads) within the thickened tissues on the tip of the irregular surface of the medial talar process

Full size image

There was increased tenosynovial fluid around flexor hallucis longus in the posteromedial (n=4) and posterolateral (n=2) groups associated with tibiotalar joint effusion. Tibialis posterior fluid was present more frequently in the posteromedial (n=4) than the posterolateral (n=1) group. Flexor digitorum longus fluid was present only in the posteromedial group (n=3). Mild tendinopathy of tibialis posterior and flexor digitorum longus was present in the one athlete with the enlarged medial talar process, while another athlete in the posteromedial group had a longitudinal split of peroneus brevis.

No osteochondral injuries were present in either group. Abnormalities of the main tibiotalar ligament (loss of striation) were present only in the posteromedial group. The anterior talofibular ligament was absent in two of the athletes in the posterolateral group, while in the posteromedial group the ligament was attenuated (n=2) and thickened (n=1). The other lateral ligaments were all intact.

Ultrasound and guided injection

Ultrasound identified hypoechoic nodular posteromedial soft tissue thickening deep to tibialis posterior between the medial malleolus and talus involving the PTTL in all nine athletes (Figs. 2, 3, 4). There were some areas of increased echogenicity, but Doppler assessment did not demonstrate increased flow within the thickened tissues.

All athletes tolerated the injection procedure with no immediate or delayed complications.

Clinical follow-up

The median duration of follow-up for all athletes was 18 months (range 4–24 months). All athletes had returned to their previous level of sport within 3 weeks of injection.

Eight athletes did not experience any residual or recurrent symptoms on follow-up (median 18 months, range 8–24 months). The athlete with an enlarged medial talar process had recurrent pain at 12 weeks and underwent resection of the process at 16 weeks after injection.

Discussion

Posteromedial impingement has been described as a clinical entity only in the past 12 years, and although there is a general consensus on the clinical features the exact aetiology has not been defined [3, 7]. Athletic supination injury typically occurs during stumbling while sprinting or on landing after jumping and places a tensile force across the lateral capsular tissues. This mechanism also results in medial compression and contusion of the posteromedial capsule and posterior fibres of the tibiotalar ligament between the talus and medial malleolus. It is believed the injured tissues subsequently undergo hypertrophy and scarring to form the collagenous and fibrous meniscoid lesion which precipitates clinical impingement [3, 7]. In previous investigations evaluating this condition, presentation for surgical treatment ranged from 14 to 52 weeks after injury,and although an underlying original mechanism was not always defined, supination injury was thought most likely [3, 7, 9]. In this current study all the athletes had suffered an supination injury 3–4 weeks prior to developing posteromedial impingement, with the acute lateral symptoms resolving early on during rehabilitation. The posteromedial symptoms were the only features limiting athletic function, with no clinical evidence of mechanical instability or lateral tenderness.

Isotope bone scan, sonographic and MR imaging features have previously been presented for two groups of patients with clinical posteromedial impingement [3, 7]. Paterson et al. described increased posteromedial ankle activity on isotope bone scanning in six patients (including five athletes) who presented a mean of 52 weeks after the original injury. The authors reported that this investigation ruled out other intra-articular abnormalities; however, further abnormality was detected and treated at arthroscopy [3]. Koulouris et al. described MR imaging (n=25) and ultrasound (n=17) features in 25 patients, of whom only two were elite athletes [7]. The MR imaging findings described were loss of striation of the PTTL (n=25) as well as abnormal signal abutting and encasing the tibialis posterior (n=10), flexor digitorum longus (n=2) and flexor hallucis longus (n=4). The authors described similar sonographic features with a hypoechoic structure presumed to be a disrupted PTTL abutting the posteromedial tendons (in 12 of 17 cases) with encasement and tethering of tibialis posterior in six of these cases [7]. In the current study MR imaging showed increased signal in the posteromedial capsule and PTTL in nine cases and displacement of the tibialis posterior and flexor digitorum longus in seven, with disruption of the PTTL fibres in only four cases. The control group of symptomatic athletes without posteromedial impingement also showed posteromedial synovitis in two cases but no tendon displacement. In all cases ultrasound identified posteromedial capsular thickening deep to tibialis posterior and obscuring detail of the underlying PTTL. Therefore the main difference between the current study and previous imaging studies appears to be the relatively low incidence of PTTL disruption and tendon encasement. This may relate to the fact that all our patients were elite athletes referred for imaging within 4 weeks of the precipitating injury, while previous series have evaluated patients at a later stage (14–52 weeks) when the abnormal tissues may have undergone further organisation.

Image-guided injections into the ankle have mainly been described involving fluoroscopy-guided anaesthetic injection to ablate symptoms and aid clinical diagnosis [12]. Two papers have described therapeutic fluoroscopy-guided tenography and os trigonum injection with steroid and anaesthetic in patient groups with tenosynovitis and posterior impingement (due to an os trigonum) respectively [13, 14]. Jaffee et al. reviewed 652 ankle tenograms and believed that the therapeutic effect was due to a combination of local anaesthesia, the anti-inflammatory effect of corticosteroid and the mechanical dilatation of the tendon sheath by injectate [14]. They reported complications of skin discoloration in 14 patients, but no cases of infection. No side effects were reported in our patient group, but the site of the injected area, deep to the posteromedial tendons, may have made skin discoloration less likely.

Only two patients with posteromedial impingement from the previous series are reported as having received steroid injections. These were performed without imaging guidance and produced only temporary relief [3]. The ultrasound-guided injection technique used in the current paper may have been more effective than blind injection because the needle was placed directly into the area of soft tissue abnormality. Additionally, the athletes in this study were all evaluated relatively soon after symptom onset compared with other series. This may have resulted in less fibrosis, allowing distension and infiltration to be performed, disrupting the abnormal tissues and allowing repair to occur [14, 15].

Previous papers discussing treatment of posteromedial impingement have presented the surgical outcome for resection of collagenous tissue from the posteromedial corner without formal PTTL repair [3, 7, 9]. However, concomitant injuries were also seen, necessitating ligament and chondral repair at surgery [3, 7]. This may have led to the longer rehabilitation times seen in those studies, while the athletes in this study commenced rehabilitation 48–72 h after injection, increasing activity over the next 10–14 days and returning to pre-injury activity within 3 weeks. All the athletes in the current study had significant osteochondral injury excluded on MR imaging and mechanical instability excluded clinically, prior to treatment with steroid injection.

Limitations of this study include its retrospective nature and lack of treatment controls. It is difficult to perform a randomised trial of treatments in elite athletes where the alternative treatment would have been continued rehabilitation, potentially delaying return to competition. There was also no control for injection technique to determine whether dry needling, soft tissue disruption or steroid were the most effective components of the procedure. No attempt was made in this study to evaluate intra- or interobserver reproducibility, as all readings were made by consensus. However, this may form a basis for further study. The clinical outcome measure in this study was return to previous full athletic activity. However, hindfoot questionnaires were developed for the general population and potentially allow elite athletes to attain excellent scores for pain and function but still not be able to return to full athletic activity.

In conclusion, this study has shown that MR imaging can highlight posteromedial synovitis in athletes with subacute clinical posteromedial impingement but that these features are not specific. If significant concomitant injuries are excluded clinically and on MR imaging, ultrasound can identify capsular abnormality and guide effective injection treatment, which allows the majority of athletes to rapidly return to activity without surgical intervention.